hi folks, i know the title sounds too cool for what it is but it describes best my project... long story short i intend to charge a layden jar or self made capacitor with my tesla coil at distance trought a needle to have some (yeah ) sparks on a spark gap, the idea was to use like a DY86 vacuum tube (CRT rectifier) to have an hi voltage rectifier that may also be fast enough to cope with the ~600kHz of the tesla, seems to need some speed that not all the valves can deliver (kinda linked to anode surface?)... honestly i already tryed and no luck, not even an accidental shock and swearing

my question is::
considering that i happen to have a box of old (but working) valves from consumer TVs and radios what could do the job best? (if could)... i have some DY86\7 and U35 lookalikes, olso other "low" voltage diodes and stuff... just asking for a direction of choice or whatever suggestion

and at the end, what frequency a DY86 can reach? seems not specified

 

That is a E.H.T. rectifier the A to K capacitance is 1.7pf if that is anything to go by.
Max.

 

i maybe overlooked that, it sounds like nothing but on a "resembling" ltspice simulation i did now, adding like 3pf to a really good diode sinks the voltage to near-nothing level... hum!... maybe i need to "zap" it more strongly, causing an arc or more corona, also trying a "less self made" capacitor, if i find one with decent voltage rating in my junk collection

 

Charging a capacitor with a high frequency (AC) voltage is not something that seems possible. You need a rectifier, but for that voltage you are probably out of luck.. Most appliance tubes will arc over at a much lower voltage than what you get from a Tesla coil.
Typically the Leyden jar type capacitors were charged using a Wimhurst static generator, which provided a single polarity.

 

well, what i hooked up is basically a single diode rectifier to a capacitor, it could also be that the capacitance is so low that it may oscillate along the "waves" from the tesla, it's just 740pf

back to checking some allowable frequency on the DY86... no, not using duper voltages on the scope of course

 

Hello,

The high voltage in a television set usualy works at the line frequency.
This will be in the range of 15 to 16 kHz.

Perhaps this pages will interest you:
http://www.kronjaeger.com/hv/hv/comp/rect/index.html
https://www.jonshobbies.com/home-made-high-voltage-rectifier-strings-of-diodes.html

Bertus

 

You can get at the dump, microwave oven, HV rectifier.

Specs vary from one manufacturer to another.

Typically look something like this -

Regards, Dana.

 

At around 50 kV a DY86/87 will stop acting as a rectifier. The residual gasses inside will ionise and you will then have a conductor. Unfortunately, the materials used for the filament/cathode coatings and also the metalised gettering,anode material and even the glass all out gas to some extent.
https://www.radiomuseum.org/tubes/tube_dy86.html
Take a look at the spec sheet, 15kV max working voltage.
Another thing to consider is how you are powering the heaters, because they are connected to the cathode internally, they were usually fed from a heavily insulated winding of a couple of turns directly on the tv Line output transformer. The capacitive and magnetic coupling will add many more problems. You could try running them from a small well insulated battery mounted very close to the valve, but that may just not be practical.
I cannot remember what they where called, but there are some very special extremely high voltage thermionic rectifiers that were available in the 60's.
The diodes that Dana has shown are low frequency, high current ht rectifiers max voltage around 5kV and would be destroyed the moment you bring your Tesla coil output near them.
Try contacting http://qro-parts.com/ He often has some amazing stuff.

 

You can get at the dump, microwave oven, HV rectifier.

Specs vary from one manufacturer to another.

Typically look something like this -

Regards, Dana.

Hi Dana, those diodes work at mains line frequency, High current, high voltage up to around 5kV but would be destroyed by a Tesla coil output.

 

I stand corrected, Tesla voltages >> microwave oven diode capability. Or
their speed. I was not paying attention.

Do not use them in this application.

Regards, Dana.

 

The high voltage in a television set usualy works at the line frequency.
This will be in the range of 15 to 16 kHz.

yeah, infact it appears to not go very much beyond that

At around 50 kV a DY86/87 will stop acting as a rectifier.

i see, anyways i don't intend to go so high (unluckily), the spark gap could be for 10-20kV and the leiden is of course not a piece of art

Another thing to consider is how you are powering the heaters, because they are connected to the cathode internally, they were usually fed from a heavily insulated winding of a couple of turns directly on the tv Line output transformer. The capacitive and magnetic coupling will add many more problems. You could try running them from a small well insulated battery mounted very close to the valve, but that may just not be practical.

it is instead (i'd say), i use an AA, it settles at 1.37V so i suppose it's good, the filament is lit, the whole is on top of the jar

not gonna buy anything for just an experiment i could abandon if doesn't make any result, i bet special stuff is expensive

anyways it seems to not go beyond 500kHz of frequency, at ~1MHz the rectifying efect seems to be overwelmed by anything else wimpy that i've used (i only have wimpy stuff LOL), i arranged some kind of oscillator with a 200V mosfet powered at 100V, at 450kHz the effect is pretty visible (comparing filament on, filament off... i think it's legit), past that it's basically hard to see on the scope, the "thing" is pathethical, has a super duper internal resistance, can't hold almost any current and so on

i'll maybe try with some UF4007 in series (next in purchase list), i don't think i require so much current to melt em in reverse, it must charge just a leiden like once per second...

 

Don't waste your money on the uf4007's, there are specific problems when putting diodes in series working at high voltages. I once did resort to a similar method whilst waiting for the correct diodes to arrive, but without parallel balance resistors and the cumulative problems in physical layout to prevent arc-over and a host of other undesirable effects, it is not worth the expense or time.
Running at your frequency, they would not be suitable. They are only "Ultra fast" diodes, not R.F diodes. I've got 10,000 1N914's you could try, (sorry, just joking, I do have them, but they would still have all the problems listed above)

 

i think i will stick with the DY86 if i succed to make it work

 

Don't waste your money on the uf4007's, there are specific problems when putting diodes in series working at high voltages. I once did resort to a similar method whilst waiting for the correct diodes to arrive, but without parallel balance resistors and the cumulative problems in physical layout to prevent arc-over and a host of other undesirable effects, it is not worth the expense or time.

roger!!... infact i would use a more specific rectifier for the task

I've got 10,000 1N914's you could try, (sorry, just joking, I do have them, but they would still have all the problems listed above)

you are inviting me at literally a "knitting event" LOL... it would be easier to buy a big pack in china rather than ship from US(?) i find bottom end components fairly good from that source (but just bottom ones... guess why)

 

i think i will stick with the DY86 if i succed to make it work

View attachment 162045

Look at the capacitance, 17pF whilst 1N914 is 2-4pF, Recovery 4nsec and that is not the only problem with the uf400X range.
Stringing diodes in series becomes so problematic in you desired application. There is a tendency for one diode to fail s/c which then raises the voltage across the remaining good ones until another fails, then so on until you have total failure. It is a system used successfully in lots of equipment with appropriate balancing resistors, but usually at lower voltages and frequencies obtainable from your Tesla.
Out of interest, can you post a picture and some info on your Tesla coil?

 

Hello,

High voltage diodes will likely not have a fast recovery time.
The ones on this page even have a recovery time of 100 - 150 nS:
http://hvstuff.com/high-voltage-diodes-and-rectifiers

Bertus

 

Unless you are really wanting to use the Tesla coil, why not make a hV generator running at 20-50 kHz and use a Cockroft-Walton voltage multiplier using easily obtainable diodes and capacitors?

 

Unless you are really wanting to use the Tesla coil, why not make a hV generator running at 20-50 kHz and use a Cockroft-Walton voltage multiplier using easily obtainable diodes and capacitors?

at this point it's worth the shot, using more freq. than mains sounds "stiffer" on the capacitors, but i was looking to come up with a "defined" experiment that vould be replicated easily just having already a whatever tesla and a trashy diode, maybe posting it on YT (and mentioning all of you?, why not)

that's the tesla in question, it's my "everyday tesla coil", i have a smaller one and a biger one i'm working on (almost complete, already running), the small one couldn't run decently, the BJTs blew constantly (chinese fakes, yes, it's true) running on a PL84 valve worked like a charm, i was impressed, it's 2.7MHz, this one pictured was refined sometimes, i added litz wire and tuning capacitor on primary (had to revisit tuning, it's unstable), added topload (this way it doesn't spark but emits more field for tricks), has 20W transformer, runs on like 25V, safe to touch etc as far as i know, there are some details i wrote down in case i share it, i developed the circuit but even if it works apparently good with the IRF640 almost cold i'm not sure about that "bridging" capacitor i added between bases and collectors, maybe corrects a bit the phase, that's why it worked when i added that, runs continuous wave, little luck getting longer sparks interrupting it, in free space it can reach 3cm, 4-5cm getting something near it, no aparent troubles letting it run indefinitely... and the leyden, just the concept of it, not wasting time making it beautiful

 

collectors

*emitters

 

If I can locate, I will post you a schematic of a driver circuit I developed about 6 years ago that is almost bullet proof. It will work with a few component changes from 50 kHz to around 250 kHz, and in various configurations, we used it to drive all sorts of things including a Jacobs ladder and a very high voltage capacitor charger. It runs with a virtual short circuit on the output and is auto power limiting. One Jacobs ladder was 6 ft tall, spaced about 3/4 inch at the bottom and the arc would stop at around 1 ft spacing.
I did make several versions, some powered directly from the mains line output, 12 volt car battery and 24 volts from 2 17 Ah sla batteries.
I will only send the info to you in a private msg and will not post them directly on this site as they are a serious electrocution risk and unlike a Tesla coil, they produce enough current along with the voltage to fry you. P.M me if you are interested.
P.S. In one version that drove a bridged output circuit using 16 high power Mosfets would deliver around 3kW !